Updated 4:13 PM EST
Updated 11:06 AM EST. Alex rises to the #2 best as Caleb falls from 2nd to 4th. No change in the 3 most probable.
Initial post 12:21 AM EST. I will update once Votefair fills out a bit more tomorrow.
|M.K. Nobilette||F**kin’ Perfect||22||0||2||0.507|
|Majesty Rose||Wake Me Up!||56||4.091||5||0.338|
|Malaya Watson||When I Was Your Man||82||0||8||0.300|
|Jessica Meuse||Pumped up Kicks||59||0||11||0.247|
|Caleb Johnson||The Edge of Glory||55||0||14||0.193|
|Alex Preston||Story of My Life||75||0.667||19||0.112|
|Sam Woolf||We Are Young||56||0||20||0.110|
So look: things are bad for M.K. Nobilette. That being said, things have looked bad for a lot of contestants, and they made it anyway. Think of Lazaro Arbos last year. The internet was sure he was gone after he sang “In My Life” in the Top 9. But he wasn’t—he wasn’t even in the bottom 3. It’s not certain that she’s not-safe. It is, however, highly probable.
Dexter Roberts has never been in the bottom 3. His scores have been low on all indices since he started, and the inferences you’d make from those scores have never been correct. Tonight may be no exception, but, again, I wouldn’t count on it.
C.J. Harris was a wildcard pick. A few weeks ago I analyzed the history of wildcard picks, concluding that men usually last about 6 rounds and women 4. It’s been only 4, but C.J.’s marks have been much lower than George Huff, Matt Giraud, Stefano Langone, or Anoop Desai. You make the call. I think he’s in trouble.
People have asked why I continue to include Dialidol in the projections, suggesting perhaps that another poll might fare better, such as MJsBigBlog. Sure, that might be better. But then again maybe not. M.K., for instance, was the most likely to go home on that poll for the past two weeks, but she didn’t. Lazaro was the top pick in season 12 on the MJ poll for the Top 9 through Top 6, including one week in which he was in the Top 3 (!).
My long-winded point is that lots of polls say lots of things. You have to judge the way that a poll has done throughout past years before you can take its word into account and weigh it against other variables. I don’t have all the poll data for past years from that poll or any other. A request I sent to MJ for access to her poll data went unanswered, so if I’m going to do that, I have to put in a lot of labor to collect the results of her polls in past years. Sure, it may be worth it (I have collected the past two seasons, but finding the poll results on her site is difficult). But my guess is that the results will more or less mirror the WNTS ratings, which would mean they might not provide much new information to the model.